François Chollet @fchollet Deep learning @google. Creator of Keras, neural networks library. Author of 'Deep Learning with Python'. Opinions are my own. Oct. 18, 2019 1 min read

There's a big difference between data-efficiency and information-efficiency. We could make considerable progress on the former without even starting to approach the latter -- and never realize our fundamental mistake.

Abstraction is the golden path to information efficiency. But all kinds of hacks can get you data efficiency.

The difference between data & information:

- Data is plentiful & noisy, the large majority of data is not information
- Information is what makes data useful
- We are bad at extracting info from data. We only manage to distill & use a small fraction of what is really available

Working on "data efficiency" typically means finding new hacks to increase the amount of information we can extract from the same data. But "information efficiency" is an entirely unrelated concept: how to efficiently and effectively make use of information

You can follow @fchollet.


Tip: mention @threader_app on a Twitter thread with the keyword “compile” to get a link to it.

Enjoy Threader? Sign up.

Threader is an independent project created by only two developers. The site gets 500,000+ visits a month and our iOS Twitter client was featured as an App of the Day by Apple. Running this space is expensive and time consuming. If you find Threader useful, please consider supporting us to make it a sustainable project.